Transforming Support Staff Name Column

MycahD

Hello, ETL question.

For context, I am creating a dashboard for self-service reports on customer satisfaction surveys.

Typically I would create a drop down filter card to 'Filter by: Owner', but in this case the owner
column contains multiple names:

IE:
Owner

Joe Smith, Bob Smith

Cate Smith

Cindy Smith, Sam Smith

 

Does anyone have suggestions on how I can transform this column(s) so that it is possible to filter the dashboard by any one individual's name?

Thank you

ST_-Superman-_

Hi @MycahD - You are on the right track. Do you have a max of 2 support staff per survey? If not, make sure to include up to the max in your split columns.

The next step would be to leverage the DDX brick "Searchable Filter App (Multiple Columns)"

You can follow along with @BenSchein as he walks through how to set up the app on a live stream here:

https://www.youtube.com/live/XOZI0J0do6Y?feature=share

If you include the columns Support Staff 1, Support Staff 2, 3, … in your search config dataset, then you will be able to search for any single name.

The other option would be to unpivot your data so that you get multiple lines per survey, but a single line per survey and support staff. By doing this you would need to be careful about how you use the dataset because you will possibly be counting surveys multiple times when they have multiple owners.

ST_-Superman-_

Hi @MycahD - You are on the right track. Do you have a max of 2 support staff per survey? If not, make sure to include up to the max in your split columns.

The next step would be to leverage the DDX brick "Searchable Filter App (Multiple Columns)"

You can follow along with @BenSchein as he walks through how to set up the app on a live stream here:

https://www.youtube.com/live/XOZI0J0do6Y?feature=share

If you include the columns Support Staff 1, Support Staff 2, 3, … in your search config dataset, then you will be able to search for any single name.

The other option would be to unpivot your data so that you get multiple lines per survey, but a single line per survey and support staff. By doing this you would need to be careful about how you use the dataset because you will possibly be counting surveys multiple times when they have multiple owners.